1333150 九、發明說明： 【發明所屬之技術領域】 本發明係有關於一種伺服哭之$ … _ _ y 。。之’丁'、说官理保護裝置，更 砰而S之，係關於用以將伺服器中 留-> 分r β 弟或第二機箱管理 兀連接至祠服裔背板之系統管理保護裝置。 【先前技術】 哭上的不斷擴大’許多企業漸漸開始將飼服 籲裔正合到集中的資料中心，將分散在企業各處的多台小型 2服ί整合於—處’以降低管理分散所帶來的高成本並且 克服官理上的困難，因此便形成一種新形態之伺服器，我 們稱之爲刀鋒型伺服器（Blade Server)。該刀鋒飼服哭 具有體積小、節省空間、易於管理、成本低以及可快速架 構等優點，而且可相容於不同的運算系統。該刀鋒型饲服 器將處理器、記憶體、硬碟等整合於單一主機板上，以形 成一具有伺服器系統功能的主機板，而每一個主機板我們 •稱之爲一個伺服器葉月。這些伺服器葉片裝配於背板 (k Plane )’而且共用一個機箱（chassis )、電源供 應器（Power Supply)、鍵盤（keyb〇ard)、滑鼠（M〇use) 與捧頁示器（display)。 為了可控制資源分享，裝設二機箱管理板（Encl〇sure1333150 IX. Description of the invention: [Technical field to which the invention pertains] The present invention relates to a server crying $ ... _ _ y . . The 'Ding', said the official protection device, more simplistic, is about the system management protection used to connect the server -> the r β 弟 or the second chassis management 祠 to the 祠 背 back Device. [Prior Art] The ever-expanding of crying's many companies are gradually beginning to combine their food and clothing into a centralized data center, and integrate multiple small 2 garments scattered throughout the enterprise to reduce the dispersion of management. The high cost and overcoming the administrative difficulties, thus forming a new form of server, we call it the Blade Server. The blade feeds crying has the advantages of small size, space saving, easy management, low cost and quick construction, and is compatible with different computing systems. The blade-type feeder integrates a processor, a memory, a hard disk, etc. on a single motherboard to form a motherboard with a server system function, and each motherboard is called a server leaf. These servo blades are mounted on the backplane (k Plane) and share a chassis, power supply, keyboard (keyb〇ard), mouse (M〇use) and page display (display) ). In order to control resource sharing, install two chassis management board (Encl〇sure
Management Board ;以下簡稱為EM )於刀鋒型飼服器之 月板上，其中—EM爲主要，另一 EM為備用。在這種方式 中，假如主要EM當機或損毀時，備用EM立即轉換成為主 要EM，俾以保護系統，如此，於主要em當機或損毁時， 5 19692DP01 1333150 .仍可維持運作整個該刀鋒型伺服ϋ的作業。主要及備用 Μ刀别具有獨立之電路’而現今將兩者之電路合而為— •以節省空間。於刀鋒型祠服器中，主要及備用腿均係透 $匯流排直接連接至該背板處，當主要ΕΜ損壞時，並不 月b有效切斷其與背板匯流排的連接，而使主要及備用抑 之訊號串聯，進而造成電路誤動作與更進—步毁損該⑽ 中的晶片或電路。 φ 因此，如何提出一種新的伺服器之系統管理保護裴 置，以便於當伺服器中之主要踫發生狀況時即時將主要 及備用間之連接訊號切斷以免造成飼服器内發生訊號 短路具已成爲目前業界亟待克服之難題。 【發明内容】The Management Board (hereinafter referred to as EM) is on the moonboard of the blade-type feeding device, where -EM is the main and the other EM is the standby. In this way, if the main EM crashes or is damaged, the standby EM is immediately converted into the main EM to protect the system. Thus, when the main em crashes or is damaged, the 5 19692DP01 1333150 can still maintain the entire blade. Type servo operation. The main and standby boring tools have separate circuits' and now the two circuits are combined - to save space. In the blade type servo device, the main and spare legs are directly connected to the back plate through the bus bar. When the main port is damaged, the month b effectively cuts off the connection with the back plate bus bar, so that The main and standby signals are connected in series, which causes the circuit to malfunction and further damage the wafer or circuit in the (10). φ Therefore, how to propose a new system management protection device for the server, so as to cut off the connection signal between the primary and standby rooms when the main defects in the server occur, so as to avoid signal short circuit in the feeder It has become a difficult problem to be overcome in the industry. [Summary of the Invention]
鑒於上述習知技術之缺點，本發明之主要目的在於 供：種飼服器’其中一第一機箱管理單元與一第二機箱管 理單元二者之-係經由—匯流排連接至―背板，另—則未 與該匯流排連接，藉以當機箱管理單元發生電路動作錯誤 時，防止該第一機箱管理單元與該第二機箱管理單元間存 在干擾訊號，並且避免晶片燒毁或機箱管理單元的迴路毁 損。 又 為達上揭目的，本發明提供一種伺服器，包括：—背 板、-第-機箱管理單元以及—第二機箱管理單元。該^ ；:機箱^理單元包含—第1關單元與—第-保護控制 單7G。第一開關單元係經由該匯流排而連接至該背板，並 依據一第一開關訊號，決定是否將該第一機箱管理單元的 19692DP0] 6 1333150 訊號經由該第一開關單元而傳送至該匯流排。該第一保護 '控制單元係輸出該第一開關訊號，並根據該第一機箱管理 單元與該第二機箱管理單元的狀態，決定該第一開關訊號 • 的可行狀態。該第二機箱管理單元包含一第二開關單元以 - 及一第二保護控制單元。該第二開關單元係經由該匯流排 而連接至該背板，並依據一第二開關訊號，決定是否將該 第二機箱管理單元的訊號經由該第二開關單元而傳送至 該匯流排。該第二保護控制單元係輸出該第二開關訊號， 鲁並根據該第二機箱管理單元與該第一機箱管理單元的狀 態，而決定該第二開關訊號的可行狀態。 該第一機箱管理單元包括一第一選擇單元，用於獲取 該第一機箱管理單元的狀態且輸出一第一狀態訊息，其中 ' 該第一狀態訊息係表示該第一箱管理單元為有效狀態或 非有效狀態。 該第一保護控制單元包括一第一傳輸運算模組以及 鲁一第一操作運算模組。該第一傳輸運算模組係經由該第一 傳輸運算模組接收來自該第二保護控制單元的第一通知 訊息，俾以表示該第二機箱管理單元的狀態；該第一操作 運算模組係依據一第一默認級設定、一第一狀態訊息與一 第一通知訊息，而進行操作.與運算，以產生一第一開關訊 號，其中該第一開關訊號係表示該第一箱管理單元係為有 效狀態或非有效狀態。該第一默認級設定係利用邏輯“〇” 表示主要的機箱管理單元，並且該第一默認級設定係利用 邏輯“1”表示次要的機箱管理單元。該第一狀態訊息係利 7 19692DP01 1333150 用邏輯“0”表示有效狀態的機箱管理單元，並且該第一狀 '態訊息係利用邏輯“1”表示非有效狀態的機箱管理單元。 該第一通知訊息係利用邏輯“1”表示該第二機箱管理單元 - 的狀態為有效狀態的機箱管理單元，並且該第一通知訊息 .係利用邏輯“0”表示該第二機箱管理單元的狀態為非有效 狀態的機箱管理單元。 該第一開關訊號係利用邏輯表示該第一開關單元 已被開啟，並且該第一開關訊號係利用邏輯“1”表示該第 _ 一開關單元已被關閉。該第一保護控制單元包括一第一及 閘與一第一或閘。該第一及閘係包括一第一輸入端與一第 二輸入端，分別接收該第一默認級設定與來自該第二保護 控制裝置的第一通知訊息；該第一或閘係包括一第一輸入 • 端與一第二輸入端，分別接收該第一及閘之輸出與該第一 狀態訊息，俾以輸出該第一開關訊號，其中該第一狀態訊 息係表示該第一箱管理單元為有效狀態或非有效狀態的 Φ箱管理單元。該第二機箱管理單元進一步包括：一第二選 擇單元，用於獲取該第二機箱管理單元的狀態與輸出一第 二狀態訊息，其中該第二狀態訊息係表示該第二箱管理單 元為有效狀態或非有效狀態。 該第二保護控制單元包括一第二傳輸運算模組與一 第二操作運算模組。該第二傳輸運算模組係接收來自一該 第一保護控制單元的第二通知訊息，其中該第二通知訊息 係表示該第一機箱管理單元的狀態；以及該第二操作運算 模組係依據一第二默認級設定、一第二狀態訊息與一第二 8 19692DP01 1333150 通知訊息，而進行操作與運算，以產生一第二開關訊號， ' 其中該第二開關訊號係表示該第二箱管理單元為有效狀 態或非有效狀態。 • 該第二默認級設定係利用邏輯“0”表示主要的機箱管 . 理單元，並且該第二默認級設定係利用邏輯“1”表示次要 的機箱管理單元。 該第二狀態訊息係利用邏輯“0”表示有效狀態的機箱 管理單元，並且該第二狀態訊息係利用邏輯“1 ”表示非有 鲁效狀態的機箱管理單元。該第二通知訊息係利用邏輯“1” 表示該第一機箱管理單元的狀態為有效狀態的機箱管理 單元，並且該第二通知訊息係利用邏輯“0”表示該第一機 ' 箱管理單元的狀態為非有效狀態的機箱管理單元。該第二 '開關訊號係利用邏輯“0”表示該第二開關單元已被開啟， 並且該第二開關訊號係利用邏輯“1”表示該第二開關單元 已被關閉。 Φ 該第二保護控制單元包括一第二及閘與一第二或 閘。該第二及閘係包括一第一輸入端與一第二輸入端，分 別接收該第二默認級設定與來自該第一保護控制裝置的 該第二通知訊息；以及該第二或閘包括一第一輸入端與一 第二輸入端，分別接收該第二及閘之輸出與該第二狀態訊 息，俾以輸出該第二開關訊號，其中該第二狀態訊息係表 示該第二箱管理單元為有效狀態或非有效狀態的箱管理 單7〇。 本發明之該第一及第二箱管理單元係相互運作，俾以 9 19692DP01 1333150 第:開關單元之開啟或關閉’進而連接或中斷 此僅一λ 1理單元與該匯㈣之㈣連接。因 此僅该弟一及弟二箱營呼 M r ^ -- 〜早兀一者之一作為有效狀態的 ;二:：。因而當有意外發生時，本發明的系統管理 =裝置㈣止該第—及第二箱管理單元同時連接至該 產生的毀損。 ①又或该伺服器因短路訊號所 【實施方式】 •V 猎由特定的具體實施例說明本發明之實施方 式，所—屬技術領域中具有通常知識者可由本說明書所揭示 之内谷輕易地瞭解本發明之其他優點與功效。 第1圖係為本發明之伺服器之系統管理保護裝置的 基本架構方塊示意圖。如第i圖所示，一第一機箱管理單 元（Enclosure Management) 11〇、一第二機箱管理單元 120與祠服器的其他模組單元（圖中未示）係經由一匯流 排23而連接至-背板13〇,該匯流排23可為通用序列匯 流排（UniVersal Serial Bus;則或其他種類的匯流 排。 該系統管理保護裝置100的第一機箱管理單元11〇 包括一第一開關單元111、一第一保護控制單元112與一 苐一达擇單元113;該第二機箱管理單元12〇包括一^一 開關單元121、一第二保護控制單元12 2與一第_選擇單 元 123。 該第一開關單元111與該第二開關單元121係經由該 19692DP01 10 1333150 匯流排23而連接至該背板130。該第一開關單元⑴係 依據該第—保護控制單元112的第-開關訊號D卜決定 疋否^ 5玄弟一機相官理單元110的訊號EMS1經由該第一 開，單7L 111而傳送至該匯流排23處，其中該訊號 係表不5亥第一機鈿官理單元】丨〇的一内部迴路（圖中未 不）通過该匯流排23。同樣的，該第二開關單元121係 依據該第二保護控制單元122所輸出的第二開關訊號 D2，=疋疋否將該第二機箱管理單元丄2〇的訊號踫52經 由該第二開關單元121而傳送至該匯流排23纟，其中該 訊號EMS2係表示該第二機箱管理單元12〇的一内部迴路 (圖中未示）通過該匯流排23。於本實施例中，當開關 訊號Dl(或D2)係為邏輯“〇”時，該開關單元iu(或丨21) 係被開啟，δ開關訊號D1 (或j)2 )係為邏輯“ 1，，時，該開 關單元111 (或121)係被關閉。 該第選擇單元Π 3係用於獲取該第一機箱管理單 元^10的狀態，並且據以輸出一第一狀態訊息Α1。該第 二選擇單元丨23係用於獲取該第二機箱管理單元120的狀 心並且據以輪出一第一狀態訊息A2。該狀態訊息μ或 A2分別表示該機箱管理單元110或120係為有效狀態/ j ac1： ive)或非有效狀態（inactive)。於本實施例中， 當狀悲讯息A1或A2為邏輯“〇，，時，表示該機箱管理單元 110或1 2 0為有效狀態的機箱管理單元，而* 或剛輯“1”時，表示該機箱管理單元12。或： 有效狀態的機箱管理單元。 19692DP01 11 1333150 该卜保護控制單元112依據該第—機箱管理單元 =嶋與其他機箱管理單元（例如：該第二機箱管理 早：m)，而決定該第一開關訊號01的可行狀態。該第 制單元122依據該第二機箱管理單元_狀態 …、他機箱管理單元（例如：該第—機箱管理單元ιι〇)， 而決定該第二開關訊號D2的可行狀態'。於本實施例中， 該弟二保護控制單元122輸出—第—通知訊化至㈣ :保=單元112,其中該第-通知訊息π係表示該 第一機箱管理單元12 〇的狀離。兮筮 分缺— 耵狀怎°亥弟一保護控制單元112 二 ΤΛ:…(default level)設定 B1、該第-狀態 5孔心Α1與邊第一逋知訊息ci的邏輯肤 開關訊號Μ的等級。 以“’而顯示該第- 丸因此，於該第—默認級設定^中，該邏輯“0’，係表示 為-主要的機箱管理單元’而該邏輯“i，，係表示為一次要 的機箱管理單元。於本實施例中，該第—默認級設定B1 可被接地，俾以設定該第—機箱管理單元ug為主要的機 箱管理此外’假如—通知訊息Π係為邏輯 “Γ，，其係表示該第二機箱管理單元12〇的狀態係為一有 效狀態的機箱管理單元。再者，假如該第-通知訊息Π 係為4G ’則係表示該第二機箱管理單元m的狀態 係為-非有效狀態的機箱管理單^。因此，依據直值表 \得了解該第-保護控制單幻12。第2圖係表示本發 明弟-貫施例之該第1護控制單元ιΐ2。於該真值表工 中，利用-及閘（AND gate) 21G與—或閘⑽邮） 19692DP01 12 1333150 220，以了解該第一保護控制單元i i 2 表1 : 第一狀態 訊息Α1 5 —保護控f. 第一默認級 設定B1 丨單元112的j 第一通知訊 息C1 1值表 "'5'^ 開關 ιϊΓ 號D1 0 〇 0 U 0 0 0 — 1 一 0 0 1 ~ — 0 — 0 0 — 1 1 1 1 0 0 1 1 — 1 0 i 1 1 ---— 「 1 i-^ 1 0 1 1 ------- nn ] 一第3圖係顯示第1圖中的第-保護控制單元112之第 二實施例。該第一保護控制單元112包括一第一傳輸運算 模組310與-第一操作運算模组犯〇。該帛一操作運算： 組320經由該第一傳輸運算模組31〇而接收來自該第二保 #護控制單元122所傳送的第一通知訊息π ’俾以取得該 第二機箱管理單元120的狀態。該第一操作運算模組32〇 依據該第一默認級設定B1、該第一狀態訊息M與該第一 通知訊息C1，而進行操作與運算，俾以產生一第一開關 訊號D1。於本實施例中，該第一操作運算模組32〇係依 據布林運异式（B〇〇iean Expression) Dl=Al+BlxCl ,進 行數字運算。因此，該第一操作運算模組32〇包括一第一 及閘（first AND gate)與一第一或閘（first0Rgate)。 δ亥第一及閘的一第一輸入端與一第二輸入端分別接收該 13 19692DP01 1333150 ::默認級設定B1與該第-通知訊息C卜該第一或閑的 輸入端與-第二輸入端分別接收該第—及閘之榦 出與該第一狀態訊息A卜俾以輪出該第—開關訊號二 如第i圖所示’依據真值表2’以了解該第二保護控 C保護控制單元122具有-第二傳輪運 4組與一弟二知作運算模組（圖中未示）以及一 閘與—第二或閘’該第二傳輪運算模組與該第二操作運瞀 模組的實施方式與結構係與該第一保護控制單元112开 該第-傳輸運算模組31〇與該第—操作運算模組32〇相 同，該第二及閘與該第二或閘的實施方式與結構與該第— 及閘與該第一或閘相同’故不予以贊述。In view of the above-mentioned shortcomings of the prior art, the main purpose of the present invention is to provide a feeding device, in which a first chassis management unit and a second chassis management unit are connected to the backplane via a bus bar. In addition, the bus bar is not connected to prevent interference signals between the first chassis management unit and the second chassis management unit when the chassis management unit has a circuit operation error, and the chip is burned or the chassis management unit is avoided. The loop is damaged. In order to achieve the above, the present invention provides a server, including: a backplane, a - chassis management unit, and a second chassis management unit. The ^;: chassis unit includes - the first off unit and the - the first protection control unit 7G. The first switching unit is connected to the backplane via the bus bar, and according to a first switching signal, determines whether to transmit the 19692DP0] 6 1333150 signal of the first chassis management unit to the convergence via the first switching unit. row. The first protection 'control unit outputs the first switching signal, and determines a feasible state of the first switching signal according to the state of the first chassis management unit and the second chassis management unit. The second chassis management unit includes a second switch unit and a second protection control unit. The second switch unit is connected to the backplane via the bus bar, and determines whether to transmit the signal of the second chassis management unit to the bus bar via the second switch unit according to a second switch signal. The second protection control unit outputs the second switching signal, and determines a feasible state of the second switching signal according to the state of the second chassis management unit and the first chassis management unit. The first chassis management unit includes a first selection unit, configured to acquire a state of the first chassis management unit, and output a first status message, where the first status message indicates that the first box management unit is in an active state Or inactive state. The first protection control unit includes a first transmission operation module and a first operation operation module. The first transmission computing module receives the first notification message from the second protection control unit via the first transmission computing module, to indicate the state of the second chassis management unit; the first operation computing module is Performing an operation and an operation to generate a first switching signal according to a first default level setting, a first status message, and a first notification message, wherein the first switching signal indicates the first box management unit Is active or inactive. The first default level setting utilizes a logical "〇" to indicate the primary chassis management unit, and the first default level setting utilizes a logical "1" to indicate a secondary chassis management unit. The first status message is 7 19692DP01 1333150 with a logical "0" indicating the active state of the chassis management unit, and the first state message uses a logical "1" to indicate the inactive state of the chassis management unit. The first notification message is a chassis management unit indicating that the state of the second chassis management unit is an active state by using a logic "1", and the first notification message indicates the second chassis management unit by using a logic "0". A chassis management unit whose status is inactive. The first switching signal is logically indicating that the first switching unit has been turned on, and the first switching signal indicates that the first switching unit has been turned off by using a logic "1". The first protection control unit includes a first AND gate and a first OR gate. The first gate system includes a first input end and a second input end, respectively receiving the first default level setting and a first notification message from the second protection control device; the first or the gate system includes a first An input terminal and a second input terminal respectively receive the output of the first gate and the first state message to output the first switch signal, wherein the first state message indicates the first box management unit The Φ box management unit is an active state or a non-active state. The second chassis management unit further includes: a second selection unit, configured to acquire a state of the second chassis management unit and output a second status message, wherein the second status message indicates that the second box management unit is valid Status or inactive status. The second protection control unit includes a second transmission operation module and a second operation operation module. The second transmission computing module receives a second notification message from a first protection control unit, wherein the second notification message indicates a status of the first chassis management unit; and the second operation computing module is based on a second default level setting, a second status message and a second 8 19692DP01 1333150 notification message, and performing an operation and operation to generate a second switching signal, wherein the second switching signal indicates the second box management The unit is either active or inactive. • The second default level setting uses a logical "0" to indicate the primary chassis management unit, and the second default level setting uses a logical "1" to indicate the secondary chassis management unit. The second status message utilizes a logical "0" to indicate the chassis management unit of the active state, and the second status message utilizes a logical "1" to indicate a chassis management unit that is not in a valid state. The second notification message uses a logic "1" to indicate that the state of the first chassis management unit is an active chassis management unit, and the second notification message uses a logical "0" to indicate the first machine's box management unit. A chassis management unit whose status is inactive. The second 'switching signal' indicates that the second switching unit has been turned on by a logic "0", and the second switching signal indicates that the second switching unit has been turned off by using a logic "1". Φ The second protection control unit includes a second AND gate and a second OR gate. The second gate system includes a first input end and a second input end respectively receiving the second default level setting and the second notification message from the first protection control device; and the second OR gate includes a second The first input end and the second input end respectively receive the output of the second AND gate and the second status message to output the second switch signal, wherein the second status message indicates the second box management unit A box management list that is either active or inactive. The first and second tank management units of the present invention operate in mutual operation, and are connected or interrupted by the opening or closing of the switch unit. Further, only one λ1 unit is connected to the (four) of the sink (four). Therefore, only the younger brother and the second brother of the camp called M r ^ -- ~ one of the early ones as a valid state; two::. Thus, when an accident occurs, the system management of the present invention = device (4) stops the first and second tank management units from simultaneously connecting to the resulting damage. 1 or the server is short-circuited. [Embodiment] V-hunting describes an embodiment of the present invention by a specific embodiment, and those having ordinary knowledge in the technical field can easily be removed by the inner valley disclosed in the present specification. Other advantages and effects of the present invention are understood. BRIEF DESCRIPTION OF THE DRAWINGS Figure 1 is a block diagram showing the basic architecture of a system management protection device for a server of the present invention. As shown in FIG. 1, a first chassis management unit (Enclosure Management) 11A, a second chassis management unit 120, and other module units (not shown) of the server are connected via a bus bar 23. To the backplane 13A, the busbar 23 can be a universal serial bus (UniVersal Serial Bus; or other types of busbars. The first chassis management unit 11 of the system management protection device 100 includes a first switching unit 111. A first protection control unit 112 and a first selection unit 113. The second chassis management unit 12 includes a switch unit 121, a second protection control unit 12 2 and a first_select unit 123. The first switch unit 111 and the second switch unit 121 are connected to the backplane 130 via the 19692 DP01 10 1333150 bus bar 23. The first switch unit (1) is based on the first switch signal of the first protection control unit 112. Db decides whether or not the signal EMS1 of the first phase of the system is transmitted to the busbar 23 via the first opening, the single 7L 111, wherein the signal is not the first Unit of 丨〇 The second switching unit 121 is configured according to the second switching signal D2 output by the second protection control unit 122, and the second chassis management is performed according to the second switching unit 121. The signal 踫52 of the unit 丄2〇 is transmitted to the bus bar 23纟 via the second switch unit 121, wherein the signal EMS2 indicates that an internal circuit (not shown) of the second chassis management unit 12〇 passes through the The bus bar 23. In this embodiment, when the switching signal D1 (or D2) is logic "〇", the switching unit iu (or 丨 21) is turned on, and the δ switching signal D1 (or j) 2) When the logic is 1, 1, the switch unit 111 (or 121) is turned off. The first selection unit Π 3 is used to acquire the state of the first chassis management unit 10, and output a first status message accordingly. The second selection unit 丨23 is configured to acquire the centroid of the second chassis management unit 120 and accordingly rotate a first status message A2. The status message μ or A2 represents the chassis management unit 110 or 120, respectively. Is valid ( j ac1: ive) or inactive (inactive). In this embodiment, when the sorrow message A1 or A2 is logical “〇, when the chassis management unit 110 or 120 is an effective state of the chassis management unit, and * or just “1”, the Chassis management unit 12. Or: A chassis management unit in a valid state. 19692DP01 11 1333150 The protection control unit 112 determines the feasible state of the first switching signal 01 according to the first chassis management unit=嶋 and other chassis management units (for example, the second chassis management is early: m). The first unit 122 determines the feasible state of the second switching signal D2 according to the second chassis management unit_state ... and the chassis management unit (for example, the first chassis management unit ιι). In this embodiment, the second protection control unit 122 outputs a first-to-notification to (4): security=unit 112, wherein the first notification message π indicates the separation of the first chassis management unit 12.兮筮 — — 耵 耵 亥 亥 亥 亥 亥 亥 亥 亥 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护 保护grade. Therefore, the '-the first shot is set to ^, the default value is set to ^, the logical '0' is expressed as - the main chassis management unit' and the logic "i," is expressed as one time. The chassis management unit. In this embodiment, the first default level setting B1 can be grounded, and the first chassis management unit ug is set as the main chassis management, and if the notification message is logical, It means that the state of the second chassis management unit 12 is an effective state of the chassis management unit. Furthermore, if the first notification message is 4G', it indicates that the state of the second chassis management unit m is a non-active state chassis management unit. Therefore, according to the straight value table \ to understand the first - protection control single magic 12 . Fig. 2 is a view showing the first guard control unit ιΐ2 of the present invention. In the truth table, use AND gate 21G and - or gate (10) mail 19692DP01 12 1333150 220 to understand the first protection control unit ii 2 Table 1: First status message Α 1 5 - Protection Control f. First default level setting B1 j Unit 112 j First notification message C1 1 Value table " '5'^ Switch ιϊΓ D1 0 〇0 U 0 0 0 — 1 A 0 0 1 ~ — 0 — 0 0 — 1 1 1 1 0 0 1 1 — 1 0 i 1 1 ---— " 1 i-^ 1 0 1 1 ------- nn ] Figure 3 shows the first picture in Figure 1. a second embodiment of the protection control unit 112. The first protection control unit 112 includes a first transmission operation module 310 and a first operation operation module. The first operation operation: the group 320 is via the first The transmission operation module 31 receives the first notification message π '俾 transmitted from the second protection control unit 122 to obtain the state of the second chassis management unit 120. The first operation computing module 32 The first default level setting B1, the first status message M and the first notification message C1 are operated and operated to generate a first switching signal D1. In the embodiment, the first operation operation module 32 performs a digital operation according to the B〇〇iean Expression Dl=Al+BlxCl. Therefore, the first operation operation module 32 The first and second gates (first AND gate) and a first gate (first0Rgate) are received. The first input terminal and the second input terminal of the first gate and the first gate of the gate are respectively received by the 13 19692DP01 1333150::default level Setting B1 and the first notification message C, the first or idle input terminal and the second input terminal respectively receive the first and the gates and the first state message A to turn the first switch The signal 2 is as shown in the figure i, according to the truth table 2, to understand that the second protection control C protection control unit 122 has a second transmission and a second operation module (not shown in the figure) And the first and second gates of the second and second operation modules and the second operation module and the first protection control unit 112 31〇 is the same as the first operation operation module 32〇, and the second and the second or the second implementation The same as the first AND gate or gates' which is not like to be described later - the second.
根據表1與表2,當該系統管理保護裝置刚 該第一機箱管理單元U0係為有效狀態（亦即噹第一'狀1 訊息Μ係為邏輯“〇，，，並且該第二通知訊息㈣為邏‘ 19692DP01 14 u=:該::二理保護裝置1。“貞測出該第二機箱 係為邏輯(亦即該第二狀態訊.“2 气第一門關通知訊息C1係為邏輯“〇”時）， 人#關就M係為邏輯“0，,且該第Hflf 為邏輯‘4,，。同時， /弟-開關《的係 單元ill…^ ”苐開關汛號D1控制該第-開關 敔該第一開關單元ui，俾以使得該第一 板：1〇經由該第一開關單元111而連接至該背According to Table 1 and Table 2, when the system management protection device is in the active state of the first chassis management unit U0 (that is, when the first '1 message is logical', and the second notification message (4) For the logic ' 19692DP01 14 u=: The:: The second protection device 1. "The second chassis is logically detected (that is, the second status message. "2 gas first door notification message C1 is When the logic "〇"), the person #关为M system is logic "0, and the Hflf is logic '4,. At the same time, / brother-switch "system unit ill...^" 苐 switch nickname D1 control The first switch 敔, the first switch unit ui, so that the first board: 1 连接 is connected to the back via the first switch unit 111
>m非23。以該第二開關訊號D2控制該第二開 關早兀121，並關閉該黛_ „明„ J 二機箱早元121，俾切斷得該第 、S早7G與該背板130的匯流排23間的連接。 單元管理保護裝置1〇0伯測出該第-機箱管理 ；^ ^ ® ( ^ ^ ^ m 4 Ai ^ ^ ^ 备 该弟二通知訊息C2係為邏輯“0，，時），並且, =管理：護裝置100偵測出該第二機箱管理單元m 為有效狀恶（亦即該筮-此能 丨多弟一狀恶訊息A2係為邏輯‘‘〇，，，並且 訊息Π係為邏輯“1，，時)，該第-開關訊號Μ 知為邂~ 1，且該第二開關訊號〇2係為邏輯“〇”。同時， :第二開關訊號D2控制該第二開關單元⑵，並開啟該 第-開關早兀121，俾以使得該第二機箱管理單元购 由該弟二開關單元121而連接至該背板130的匯流排 ㈡。該第-開關訊號D1控制該第一開關單元⑴，並關 閉該第-開關單元⑴，俾以切斷該第一機箱管理單元 11 〇與該背板130的匯流排23間的連接。 當該第-機箱管理單元110的該第一狀態訊息ai係 19692DP01 15 1333150 為邏輯“0，，與該第一通知訊息Cl係為邏輯“〗，，時，由、、 可知，該第-機箱管理單元110與該苐二機箱 120二者係為有效狀態，此時，該第—保護控制單元= 係依據該第-默認級設定B1，而將該第一開關 換至邏輯“〇，，；以及該第二保護控制單元122係依 刀 —默認級設定B2(例如、顯示該第二開關訊號D2 輯‘‘1’，），而將該第二開關訊號D2切換至邏輯“丨，，。同日士， 以該第-開關訊號D1控制該第—開關單元lu，並開 該第-開關單元1U。以該第二開關訊號⑽所控制該 二開關單元12卜並關閉該第二開關單元121。因此，告 該第-機箱管理單幻1G與該第二機箱管理單以2〇二^ 係為有效狀㈣’本實施例解決伺❹故障之缺點，並且 進步避免該飼服器無法有效運作之問題；避免該饲服哭 的内部迴路燒毁與該他器毁損，以及避免機箱管理單原 2短路Λ號而無任何保護裝置而連接至該背板的匯流 排。 此外，如矣1 ( 一 1所不，富該第一機箱管理單元110的第 “n,主心A1係為邏輯“ 1，，且該第-通知訊息C1係邏輯 單元122亦:、5亥第—機箱管理單元110與該第二保護控制 pi m % I為非有效狀態），該第一開關訊號D1與該第一 開關§孔5虎D2皆盔-土。 筮一 „ „ „ 马礙輯“1，，，因而該第一開關單元111與該 弟一開關早元]? 單元11〇 白為關閉狀態。因此’該第-機箱管理 排23未相互連:機箱管理單元120與該背板130的匯流 坎’這表示該伺服器係為異常的。同時， 16 19692DP01 1333150 .伺服益發出警示訊號至伺服器維修者處，使其得維修該伺 服器，所以該伺服器得以穩定地與可靠地運作，而這部分 為砘知此領域之人士所熟知的技術，在此不再贅述。 坏合上述，上述之實施例的系統管理保護裝置1 中主要係經由該苐一及弟二保護控制單元112、12 2彼 此運作而控制該第一及第二開關單元m、121的開啟或 關閉，因此，經由與其相對應的該第一開關單元U1或第 鲁=開關單元12卜該第一機箱管理單元11〇或該第二機箱 盲理單元12 0係係連接至該背板13 〇的匯流排2 3。故， 經由上述之實施例，僅有該第一機箱管理單元丨1〇及第二 機箱管理單元120其中之一經由與相對應的第一開關單 元111或第二開關單元121而連接至該背板i 3〇的匯流排 23，所以該第一及第二機箱管理單元11〇、12〇不會同時 ，接至該背板13〇的匯流排23。另外，如上述之實施例， t該第一機箱管理單元11〇或第二機箱管理單元12〇異常 _ 0寸，該異常的機箱管理單元係被阻擋連接至該背板 的匯流排23，因而機箱管理單元不會遭受干擾與來自該 月板130的匯流排23之短路訊號。習知技藝因為該干擾 與該短路訊號而造成該伺服器的内部迴路短路，並且使其 燒毁，而且該伺服器有可能毁損與無法穩定地運作；而本 發明提供的系統管理保護裝置便可克服習知技藝之此一 缺失。 惟以上所述之具體實施例，僅係用以例釋本發明之特 點及功效，而非用以限定本發明之可實施範疇，在未脫離 19692DP01 17 1333150 本發明上揭之精神與技術簕略 . — τ/、议彳ίΤ靶可下，任何運用本發明所揭示 内容而完成之等效改變及修飾，均仍應為下述之申請專利 範圍所涵蓋。 【圖式簡單說明】 -第1圖為一方塊示意圖，其顯示本發明第一實施例之 系統管理保護裝置的基本架構； 第2圖係根據第一實施例，顯示弟1圖中之第一保護 控制單元之示意圖；以及 φ 第3圖係根據第一實施例，顯示第1圖中之第一保護 控制單元之示意圖。 【主要元件符號說明】 100 系統管理保護裝置 110 第一機箱管理單元 111 第一開關單元 112 第一保護控制單元 113 第一選擇單元 120 第一機箱管理單元 121 第二開關單元 122 第一保5蔓控制單元 123 第二選擇單元 130 背板 210 及閘 220 或閘 23 匯流排 310 第一傳輸運算模組 320 第一操作運算模組 18 19692DP01>m is not 23. Controlling the second switch early 121 by the second switching signal D2, and closing the 黛 _ „ J J chassis early 121, 俾 cutting the first, S early 7G and the busbar 23 of the backplane 130 The connection between the two. The unit management protection device detects the first-chassis management; ^ ^ ® ( ^ ^ ^ m 4 Ai ^ ^ ^ prepares the second notification message C2 to be logical "0,, hour", and, = management The protection device 100 detects that the second chassis management unit m is valid (that is, the 筮-this can be a sneak peek message A2 is logical ''〇,, and the message is logical ” 1, time), the first switching signal Μ is 邂~ 1, and the second switching signal 〇2 is logic "〇". At the same time, the second switch signal D2 controls the second switch unit (2), and the first switch switch 121 is turned on, so that the second chassis management unit is connected to the backplane 130 by the second switch unit 121. Bus (2). The first switching signal D1 controls the first switching unit (1) and closes the first switching unit (1) to close the connection between the first chassis management unit 11 and the bus bar 23 of the backplane 130. When the first status message ai of the first chassis management unit 110 is a logical "0", and the first notification message C1 is a logical "", the first chassis is determined to be logically "". The management unit 110 and the second chassis 120 are in an active state. At this time, the first protection control unit = according to the first default level setting B1, and the first switch is switched to logic "〇,,; And the second protection control unit 122 switches the second switching signal D2 to logic "丨," according to the knife-default level setting B2 (for example, displaying the second switching signal D2 series ''1'). In the same day, the first switch unit lu1 is controlled by the first switch signal D1, and the first switch unit 1U is turned on. The two switching units 12 are controlled by the second switching signal (10) and the second switching unit 121 is turned off. Therefore, it is reported that the first chassis management single magic 1G and the second chassis management management table are effective. (4) The present embodiment solves the shortcomings of the servo failure, and progress prevents the feeding device from being effectively operated. Problem; avoiding the inner loop of the feeding suit crying and the damage of the other device, and avoiding the chassis management unit 2 short circuit nickname without any protection device connected to the busbar of the backboard. In addition, if 矣1 (1), the first "n, the main center A1 of the first chassis management unit 110 is logical "1, and the first notification message C1 is also a logical unit 122: The first chassis management unit 110 and the second protection control pi m % I are in an inactive state, and the first switch signal D1 and the first switch § hole 5 and the tiger D2 are both helmet-earth.筮 „ „ „ 碍 辑 “ “ “ , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , Therefore, the first chassis management row 23 is not connected to each other: the sink of the chassis management unit 120 and the backplane 130 indicates that the server is abnormal. At the same time, 16 19692DP01 1333150. The servo sends a warning signal to the server maintainer to repair the server, so the server can operate stably and reliably, and this part is well known to those skilled in the art. The technology is not repeated here. In the above, the system management protection device 1 of the above embodiment mainly controls the opening and closing of the first and second switching units m, 121 via the operation of the first and second protection control units 112, 12 2 to each other. Therefore, the first switch unit U1 or the second switch unit 12 or the second chassis blind unit 120 is connected to the backplane 13 via the first switch unit U1 or the second switch unit 12 corresponding thereto. Bus 2 2 3. Therefore, through the above embodiment, only one of the first chassis management unit 丨1〇 and the second chassis management unit 120 is connected to the back via the corresponding first switch unit 111 or second switch unit 121. The busbars of the board i 3〇 are so that the first and second chassis management units 11〇, 12〇 are not simultaneously connected to the busbars 23 of the backplane 13〇. In addition, as in the above embodiment, the first chassis management unit 11 or the second chassis management unit 12 is abnormal, and the abnormal chassis management unit is blocked from the bus bar 23 connected to the backplane. The chassis management unit does not suffer from short-circuit signals from the busbar 23 from the moonplane 130. The conventional technique causes the internal loop of the server to be short-circuited due to the interference and the short-circuit signal, and burns it, and the server may be damaged and cannot operate stably; and the system management protection device provided by the present invention can Overcome this lack of conventional skills. However, the specific embodiments described above are merely used to illustrate the features and functions of the present invention, and are not intended to limit the scope of the present invention, and the spirit and technical strategy disclosed in the present invention is not departing from the invention. — τ / 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 。 BRIEF DESCRIPTION OF THE DRAWINGS - Figure 1 is a block diagram showing the basic architecture of a system management protection device according to a first embodiment of the present invention; and Figure 2 is a first embodiment showing the first of the drawings Schematic diagram of the protection control unit; and φ Fig. 3 is a schematic diagram showing the first protection control unit in Fig. 1 according to the first embodiment. [Main component symbol description] 100 system management protection device 110 first chassis management unit 111 first switch unit 112 first protection control unit 113 first selection unit 120 first chassis management unit 121 second switch unit 122 first protection 5 vine Control unit 123 second selection unit 130 backplane 210 and gate 220 or gate 23 busbar 310 first transmission operation module 320 first operation operation module 18 19692DP01